For too long, I’ve been sitting on the banks, watching a stream of ideas flow through me. I’ve toyed with many of them; pondered, tinkered, and even brought a few close to completion; but I’ve always stepped back before they could fully take shape. None have crossed the finish line.
Now, I’m ready to commit. I’ve chosen an idea that feels revolutionary, one I suspect many others are already chasing. It burst into the spotlight about a year ago when Devin’s demo video took the internet by storm. Since then, a wave of similar tools - like Lovable, Windsurf, Cursor and many more have emerged, some of which I use regularly. I’ve been impressed, some more than others, but I still feel the dream of a true AI Software Engineer remains out of reach. It’s a concept that excites people, and I want to take my shot at it. So, starting now, I’m diving into building my own AI Software Engineer.
Let me lay out my vision clearly:
This is a massive undertaking, and I know it’ll be tough. But that challenge sparks a thrilling mix of joy and trepidation, like standing on the edge of a cliff, heart racing, knowing the leap could change everything. When I use tools like the ones I mentioned, I can’t shake the feeling they could be better. This is my chance to stop letting ideas slip away, to wade into the river and build something real.